Cholestérol haut, cholestérol bas: quels sont les risques en gériatrie ?

(1)

Cholesterol, CV risk

&

statins

in Older Persons (OP: 75+)

(an evidence-based approach)

Prof. Benoit Boland

Geriatric Medicine,

UCL-Brussels

Géri-Liège, 20 03 2015

CVRisk & Cholesterol management

in the Elderly (75+)

In the elderly, pyramid (4 faces)

1. Ageing & Health

2. Ageing & Morbi-mortality

3. Ageing & CVRisk

4. Ageing & Lipids

Complex decision making (at all ages !)

(2)

1. Older age & Health

GOALS in OP

To decrease consequences of diseases

To prolong independence

To prevent social isolation

MEANS

↑

nutrition

(

weight)

↑

physical activity

(

muscle)

↑

interactions

(

social life)

± ↓ cholesterol ?

(

? reason ?)

2. Older age & Morbi-mortality

First cause of morbi-mortality = CVD

C²VD (Coronaro&Cerebro Vascular Diseases)

Burden : non-C²VD > C²VD

Alzheimer, cancers, infections, …

Medical decisions (EBM) in older persons

Life expectancy ?

« Quality of life » ?

(3)

3. Older age & CVRisk

CVRisk : ↑ with age (exponential)

CVRisk definition = absolute 10 years risk of acute event (MI or

stroke; fatal or not)

Coronary disease : very frequent in OP

observed in 70% of 70+ (post-mortem analyses)

men > women

Gender difference (~10 yrs)

half of longevity differential

sex ratio: progressive ↑ with age

persons 80+ : 3 w / 1 m

4. Older age & Lipids

Blood changes with ageing

↓ Total Chol. & Chol-LDL

↔ HDL-Chol

Effects of diseases

Acute: inflammation

Chronic: malnutrition

Effects of statins

Association Cholesterol

CVD

(4)

B BOLAND, PCV & PA 75+ Macro-phages Stries

lipidiques

Epaissis sement Plaque d’ athérome Sténose luminale sclérose

0 – 25 ans

25 – 50 ans

50 à 75 ans

bouillie thrombose

ATHéro matose –> thrombose –> sclérose…

1

Science

Physiopathologie « fondamentale »

Observations

Connaissances théoriques

Avis d’experts internationaux

Autorité : guidelines

(5)

ScienceBM

vs.

EvidenceBM :

Physiopathologie

« fondamentale »

Observations

non-systématiques

Connaissances

théoriques

Avis d’experts

Autorité acceptée

Physiopathologie

« insuffisante »

Observations

systématiques

Démonstrations

concrètes (preuves)

Evaluation des données

Autorité contestée

EBM & hiérarchie

des différents types d’études cliniques

Niveau de

preuve

Contrôle

sur

le champ

d’observation

Etudes

d’intervention

Etudes de cohorte

Etudes cas-témoins

Etudes transversales

(6)

Two central geriatric questions …

1. Is

cholesterol a CV risk factor

at age ≥

75 years ?

2. How much is a

statin appropriate

at

age ≥ 75 years ?

clinical benefit vs. harm

evidence for links

in OP (75+) ?

high

cohorts ?

Cholesterol

CVDiseases

(yes)

RCTs ?

(7)

1. In older persons (75+),

is

cholesterol a CV Risk Factor ..?

Classical approach

(physiopathology)

EBM approach

(etiology)

Question:

« P E O » (causative arguments)

Search for valid sources of information :

Hierarchy: (SR) > cohorts > RCT_control arm

Critical appraisal

PROSPER, Lancet 2002

70+

CV HEALTH study, JAGS 2004

65+

LEIDEN study, Lancet 1997

85+

Cholesterol in

OP70+

is not a CV risk factor

(MI or Stroke)

PROSPER study

(RCT, control arm)

Lancet 2002; 360: 1623-1630

Control arm, ~2.900 persons aged 70 – 82 yrs

mean age 75 yrs; with ou w/o CV event

Events : 3.2 yrs of f/up

MI: NO association with LDL-C, but with HDL-C

Stroke: weak association with Chol.

CV Mortality : NO association with Chol.

(8)

Cohort study : design

OP without CVD

Cholesterol +

(« high »)

Cholesterol

-(« low »)

Time

follow-up follow-up

t

₀

t

1

CVevent +

CVevent

-CVevent +

CVevent

-Cholesterol in

65+

is not a CV risk factor

(MI or Stroke)

CV HEALTH study

(cohort, 7.5 yrs of f/up)

JAGS 2004; 52: 1639-47

4.885 persons 65+ (mean 73 yrs), free of CVD

Events : 436 MIs, 332 Strokes, 1096 deaths

MInfarct : NO association with Chol.

Stroke : weak association with Chol.

CV Mortality : NO association with Chol.

(9)

Cholesterol in

OP85+

is not a factor for CV risk or death

LEIDEN study

(Cohort, 10 yrs: 1987–1996)

Lancet 1997; 350: 1119-1123

All inhabitants aged > 85+ (born 1883–1901)

n=750; mean age 89 yrs; ~90% free of CVD at baseline

Mortality: 10 yrs of f/up

high : 90% at 10 years

elevated cholesterol (260 mg/dl)

is not a RF for CV mortality

is a PF for non CV mortality (infection, cancer)

(10)

BASELINE CHARACTERISTICS, year 1987

Weverling-Rijnsburger AW, Lancet 1997;350:1119-23

Subjects (no)

Mortality risk

Unadjusted Adjusted age and sex

Adjusted age, sex and

≥≥≥≥6.5 mmol/L 171 0.56 (0.45-0.69) 0.62 (0.49-0.77) 0.64 (0.50-0.82) 5.0- 6.4 mmol/L 350 0.72 (0.60-0.86) 0.78 (0.65-0.94) 0.81 (0.66-1.01)

< 5.0 mmol/L 203 1.00 1.00 1.00

10 - year mortality risks adjusted for various determinants

Total cholesterol

(11)

All cause mortality

0

2

4

6

8

10

0

0.2

0.4

0.6

0.8

1 E

st

im

a

te

d

M

o

rt

a

lit

y

P

ro

b

a

b

ili

ty

Years of follow-up

< 5 mM;

< 200

5.0-6.4 mM; 200-259

≥≥≥≥

6.5 mM ;

> 260

Lancet 1997;350:1119-23

All cause mortality

0

2

4

6

8

10

0

0.2

0.4

0.6

0.8

1 E

st

im

a

te

d

M

o

rt

a

lit

y

P

ro

b

a

b

ili

ty

Years of follow-up

< 5 mM;

< 200

5.0-6.4 mM; 200-259

≥≥≥≥

6.5 mM ;

> 260

(12)

Cardiovascular mortality

0

2

4

6

8

10

0

0.2

0.4

0.6

0.8

1 E

st

im

a

te

d

M

o

rt

a

lit

y

P

ro

b

a

b

ili

ty

Years of follow-up

< 5 mM;

< 200

5.0-6.4 mM; 200-259

≥≥≥≥

6.5 mmol/L; > 260

Lancet 1997;350:1119-23

Infectious mortality

0

2

4

6

8

10

0

0.1

0.2

0.3

0.4

0.5 E

st

im

a

te

d

M

o

rt

a

lit

y

P

ro

b

a

b

ili

ty

< 5 mm ; < 200

5-6.4 mM; 200-259

≥≥≥≥

6.5 mM ; > 260

Years of follow-up

(13)

Very Old Persons (85+) & cholesterol

Conclusions

• High mortality risk (90% at 10 years)

• CVD is the first cause of mortality

• total chol. is not a risk factor for CV death

• high total chol.

increased survival !

+ 1 mM chol =

-

15 % in mortality (RR 0.85 ; 0.79-0.91)

• low total chol.

increased risk of death

due to infection or cancer or …

Lancet 1997;350:1119-23

(14)

Serum total cholesterol and 10-year mortality in an 85-year-old Japanese population

Introduction. Little is known about the association between total cholesterol (TC) and cause mortality in the elderly. Here we examined the association between TC and all-cause mortality in 207 very elderly (85-year-old) participants.

Methods. At baseline in 2003, collection of laboratory blood tests, blood pressure (BP) and body mass index (BMI), and lifestyle questionnaires. Follow-up during the subsequent 10 years. In 2013, of the 207 participants in 2003, 70 participants had survived, 120 individuals had died, and 17 were lost to follow up. The TC values were divided into high-TC (≥ 209 mg/dL), intermediate-TC (176–208 mg/dL), and low-TC (≤ 175 mg/dL) categories. Kaplan–Meier method for survival analysis.

Results. Both the high-TC and intermediate-TC groups survived longer than the low-TC one. The men with high TC survived longer than those with low TC, but no corresponding difference was found for the women. A multivariate Cox proportional hazards regression model, with adjustment for gender, smoking, alcohol intake, history of stroke or heart disease, serum albumin concentration, BMI, and systolic BP, revealed that the total mortality in the low-TC group was 1.7-fold higher than that in the high-TC group. Adjusted for the same factors, mortality decreased 0.9% with each 1 mg/dL increase

in the serum TC concentration and decreased 0.8% with each 1 mg/dL increase in the

serum (low-density lipoprotein) LDL-cholesterol (LDL-C)

These findings suggest that low TC and low LDL-C may be independent predictors of high mortality in the very elderly.

evidence for links

in OP (75+) ?

high

NO

Cholesterol

CVDiseases

(yes)

RCTs ?

(15)

2. In OPersons (75+),

how much is a

statin appropriate ..?

Classical approach

(generalisation from < 75 !)

EBM approach

(efficacy & safety in 75+ : ?)

Question « P I/C O »

Search for valid sources of information:

Hierarchy: (Guidelines) > (MA) > RCT (few)

Critical appraisal

HPStudy, 2002

40-80 yrs

70+: n=5.806

PROSPER, 2002

70-82 yrs

70+: all : n=5.804

JUPITER, 2008

70-80 yrs

70+: all: n+5.695

3 voies de réponses, selon l’EBM

Fort peu de données chez OP

1. Etudes randomisées classiques

A. HPS_2002

A. PROSPER_2002

B. JUPITER_2010

2. Etude randomisée de terrain

3. Consensus STOPP/START

(16)

HPS, 2002

Heart Protection Study

P

20.536 patients, 40-80 yrs

28 % > 70 yrs :

5.806 were EP70+

(post-hoc)

at high CV risk : SPrevention, or Db2+other RF(s)

I / C

Simvastatin 40 vs. placebo, 5 years

O

CV morbi-mortality (nf MI, Sk, CV†)

↓ 25 %

(p < 0,001)

Comments

Statin is effective in 70-80 years at high risk (SPrevention or Db2)

with no differences related to gender, age (> 70 yrs), nor LDL-C !

MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20.536 high-risk individuals: a randomised placebo-controlled trial. Lancet 2002;360:7

PROSPER, 2002

Prospective Study of Pravastatin in the Elderly at Risk

P

5.804 pts (51% women),

aged 70-82 yrs

mean 75 yrs

all were EP70+ (n=5.804)

SPrev.

(n=2500)

or High risk PPrev.

(n=3300)

I / C

40 mg pravastatin 40 vs. placebo, 3.2 years

O:

CV morbi-morta (nf MI, Sk, CV†)

↓ 15% overall

(p=0,01)

CAD deaths ↓ ; non-fatal MI ↓ ; new cancer ↑

Comments (no effect on MMSE)

Not effective in Primary Prevention

+ in SPrev.

nor in women

+ in men

Little effect if HDL-C > 50

+ if low HDL-C

(17)

Statines et 4ème âge, BBoland

PROSPER

[Lancet 2002;360:1623]

P:

5.804 pts, âge 70-82 ans (µ:75 ans; ♀ 51%)

HRCV, en PCV2

nd

_(57%)

ou en PCV1

re

_(43%)

I/C: pravastatine 40 vs. placebo, 3.2 ans

O:

Evénement

Fréquences/an

RRR (p-value)

NST/an

IM, AVC, †CV

16.2 vs. 14.1

15% (0,01)

48 PCV2

nd

_{21.7 vs. 17.4}

_{22% (SS)}

₂₃

PCV1

re

_{12.1 vs. 11.4}

_{6% (NS)}

_“143“

mes commentaires: une statine chez les PAgées 70+

en PCV 2

nd

_{avec haut risque CV (22%/an) est utile [surtout chez ♂ , HDL-C<45]}

en PCV 1

re

_{avec haut risque CV (12%/an) est inutile}

JUPITER

Analyse des 70+ [Ann Int Med 2010; 152: 488-96]

P: 5.695 PA en PCV

1

re

LDL<130 et CRP>0.2 (µ:74 ans; ♀ 52%) I/C: rosuvastatine 20 vs. placebo, ~ 4 ans

O: Evénement Fréquences/an RRR (p-value) NST/an IM, AVC, †tot 2.11 vs. 3.04 % 30% (0.001) 107

IMyocarde 0.27 vs. 0.50 % 45% (0.05) 437 AVCérébral 0.35 vs. 0.64 % 45% (0.02) 344 †CV 0.34 vs. 0.41% 27% (0.53) x

IM, AVC, †CV 0.96 vs. 1.55 % 38% “169“

mes commentaires: une statine chez des PAgées 70+

en PCV 1re_{avec risque CV moyen (1.5 %/an) n‘est utile que...}

(18)

B BOLAND, PCV & PA 75+

Etude randomisée de terrain, DEBATE

(Finlande)

[Am Heart J, 2006; 152: 585-92 ]

PCV 2

nd

_{chez des patients 75+ après infarctus}

Patients: n=400, entre 75 et 90 ans

Antécédent d’infarctus

Recrutés dans la population générale

Intervention: tous les ttts recommandés

Contrôle: ttts usuels

Outcomes:

Atteinte des cibles :

OK (TAs, LDL-C, …)

Diminution des acc.CV:

KO

Décès non-modifiés (18 vs. 17%) à 3.4 ans !

Délai non-modifié jusqu’à l’accident CV !

Statins: clinical efficacy

according to age (decades)

HPS others JUPITER

HPS

PROSPER

JUPITER

(-)

60 – 69 yrs

_{70 – 79 yrs}

_{> 80 yrs}

Secondary

Prevention

+ ,

+

,

+

?

Primary

Prevention

+ ,

+

_--

_,

±

?

(19)

THoMessage =

in OP 75+, evidence for links

high

NO

cholesterol

CVDiseases

NO (PPr)

YES

YES (SPr)

statins

Consensus

STOPP & START

(Prescription inappropriée)

Besoin d’un outil pertinent, pratique (rapide, simple), à jour

Initiative en Irlande :

2003: Situations fréquentes et importantes chez les PAgées

2004: Liste initiale, étude pilote

2006: Consensus, 18 experts

gériatres, médecins généralistes, neuropsychiatres, pharmaciens

2008 : 65 & 22 critères retenus, et publiés

STOPP: Screening Tool of Older People’s inappropriate Prescriptions

START: Screening Tool to Alert doctors to Right (appropriate, indicate) Treatments.

2008: études de terrain : prévalence élevée

Primary Care Hospital Adm. Nursing Home STOPP 21% 35 % 60 %

(20)

STOPP & START

Études de validation

2008: reproductibilité

2008: validation rétrospective: détection des EIM

2008: validation prospective:

Sur 600 PA, 158 (26%) ont des EIM (n=329, dont 219 ont causé ou contribué à l’Hospitalisation, parmi lesquels 150 étaient évitables dont 94 étaient détectés par STOPP)

2008: faisabilité : 2 minutes:

Sur 50 PAgées avec 418 prescriptions, 102 critères STOPP et 47 critères START

2008: RCTrial

n=400: nette diminution des médications « en trop » et en « trop peu », à la sortie, 3 et 6 mois

> 2009 : impact sur l’incidence des EI, sur les coûts, …

Version.2

…2014 : 2

ème

_{version, internationale}

> 2015 : essais cliniques européens (SENATOR, OPERAM)

Liste START

, 2014

34 omissions de prescription potentiellement inappropriée

Introduction :

prescription à envisager si indication, hors C-I et hors fin de vie

Section A: Cardio-vasculaire

Condition

Prescription

si FAuriculaire

anticoagulation (AVK ou NOAC)

& CI anticoag.

antiagrégant plaquettaire

si HTA >160/90

anti-hypertenseur

si ATHérosclérose

antiagrégant plaquettaire

statine si âge < 85 ans et EV > 1 an

si ICardiaque systo.

IEC

si ICard. systo stable

β-bloquant

si maladie coronaire

IEC

si cardioϖ ischémique

β-bloquant

(21)

Liste STOPP

, 2014

84 critères de prescription potentiellement inappropriée

Section A: Indication de prescription

A1. pas d’indication: statine/aspirine en PPrimaire, IPP sans RGO, .. A2. durée trop longue

A3. duplication de classe (BBq, IEC, AINS, BZD, …)

Section B: Cardio-vasculaire

Médicament Critères (situations à risque)

Digoxine ICard « diasto »; >125 µg/j si FGlom <50ml/min Diltiaz/Verap. ICardiaque NYHA classe III ou IV

Β-bloquant Vérap/diltiaz; FC<50; BPCO; Db2 avec hypo Amiodarone 1ère_{ligne pour tachycardie SV}

Diur. Anse 1ère_{ligne pour HTA}

Thiazide OMI périphériques, goutte, Na+, K+, Ca++ IEC/ARAII hyperK+

Aldactone IEC/ARAII, sans suivi K+

Act° centrale (sauf si intolérances aux autres classes) Vasodilatat. Hypotension orthostatique objectivée

Statine, commentaires

STOPP

START

Pas d’indication : PCV 2nd _{et « fin de vie »} PCV 1re _{(exceptions…)} Etudes: n=2 PROSPER_PP; JUPITER 410 PAgés (CUSLuc, 2010) statine: 96 / 410 (23%) versus en PCV1re_{: 47}_{/ 260}_{= STOPP}

(dont âge > 85 ans: 15) en PCV2de_{: 49 / 150}

dont âge > 85 ans : peu dont EV<1 an: ?

prescription si PCV 2de

si EVie > 1 an et si âge < 85 ans

Etude : n=1 : PROSPER_PS

âge 85±5 ans, 8± 4 médicaments/jour

pas de statine: 314 / 410 (77%)

en PCV1re_{: 213 / 260 : OK}

en PCV2de_{: 101 / 150}

dont âge < 85 ans : 57 = START

(22)

Maladies CV chez les PAgées (75+)

Processus

étiologiques

Maladies

silencieuses

Complications

cliniques

Issues

finales

Santé Fact. de Risque Age Cholestérol ? Diabète Hypertension Stase par FA Stade préclinique Maladies silencieuses Athérosclérose Artériosclérose Caillot auriculaire Stade clinique Complications Infarctus M AVCérébral Stade tardif Issues: (guérison) séquelles dépendances récidives décès

PCV & PAgées (75+)

STOPP

car « futile »

START

car « utile »

Statine PCV1re PCV2nd

& âge > 85 ans & EV<1 an

PCV2nd

& âge < 85 ans & EV>1 an

(& indépendance fct°)

anti-HTA Chute & hTAO HTA (≥ 160/90 mmHg)

Anticoagulation FA avec risque AVC <

risque HICranienne

FA avec risque AVC > risque HICranienne/décès